chore: align NL parser with project structure and docs

Author: Swaroop Manchala

nl_parser.py cortex/nl_parser.pynl_parser.py renamed to cortex/nl_parser.py

@@ -0,0 +1,32 @@
+# Natural Language Parser (NL Parser)
+
+## Overview
+The NL Parser enables users to describe installation requests in natural language
+(e.g., “install docker and kubernetes” or “set up python dev environment”).
+It converts free-form text into structured intents that Cortex can act upon.
+
+This improves demo reliability and usability by removing the need for strict
+command syntax.
+
+---
+
+## Key Features
+- Typo tolerance (e.g., kubernets → kubernetes, pyhton → python)
+- Rule-based + fuzzy semantic intent detection
+- Confidence scoring for detected intent
+- Clarification prompts for ambiguous requests
+- Slot extraction (python version, platform, packages)
+- Lightweight, dependency-free core logic
+
+---
+
+## Usage Example
+
+```python
+from cortex.nl_parser import parse_request
+
+result = parse_request("pls install pyhton 3.10 on mac")
+
+print(result)
+
+

docs/NL_PARSER.md

@@ -1,5 +1,5 @@
 import pytest
-from nl_parser import parse_request
+from cortex.nl_parser import parse_request
 
 @pytest.mark.parametrize("text,expected", [
     ("install something for machine learning", "install_ml"),
@@ -34,4 +34,6 @@ def test_corrections():
 def test_slot_extraction():
     r = parse_request("pls install python 3.10 on mac")
     assert r["slots"].get("python_version") == "3.10"
-    assert r["slots"].get("platform") in ("mac", "macos")
+    assert r["slots"].get("platform") in ("mac", "macos")
+    
+